Replicate
Run thousands of open-source AI models through one API, paying per second.
Visit Replicate Pay per use (per second or per output); free trial credits
What is Replicate?
Replicate hosts thousands of community and official open-source models — image, video, audio, and language — behind a single API, billed by the second of compute or per output. Developers use it to add a model to an app without managing GPUs.
Key features
- One API for thousands of image, video, audio, and text models
- Per-second or per-output billing with no idle cost
- Deploy your own models with Cog containers
- Fine-tune popular image models on your data
- Webhooks and streaming for long-running jobs
Pros
- Widest catalogue of open models of any hosting API
- Zero infrastructure — a model is a single HTTP call
- Pricing scales to zero when you are not using it
Cons
- Cold starts on less-popular models can take a minute or more
- Per-second pricing is hard to predict for long generations
- Community models vary widely in quality and maintenance
Best for
Developers adding image or video generation to an appPrototyping with many models before committingTeams that do not want to run GPUs
Read more
Related comparisons
Alternatives to Replicate
Fireworks AI
Production inference for open-source LLMs — function calling, structured output, fine-tuning.
FreemiumFree $1 credit, then $0.20-$5/M tokens
Released September 2022
Together AI
Fastest inference for open-source models — Llama 4, Qwen3, DeepSeek V3 at low cost.
FreemiumFree credits, then $0.20-$5/M tokens
Released June 2022
Baseten
Production inference infrastructure for deploying your own AI models.
PaidUsage-based GPU pricing; enterprise plans
Released January 2019